I collaborated with a group led by L. Biesecker (NHGRI) in identifying that mutations of the DNC gene on chromosome 17 are mutated in Amish microcephaly. This collaboration is part of a larger project entitled the Anabaptist Genealogy Database (AGDB) to construct a large computer-searchable genealogy of North American Anabaptists. I am also collaborating with the group of A Shuldiner and B. Mitchell (U. Maryland Medical School) in other usages of AGDB. I am collaborating with Dr. B. Grimbacher (U. Freiburg) to hunt genes for hyperIgE syndrome, common variable immune deficiency (CVID), neutropenia, and other disorders related to the immune system. Highlights for the past 12 months were: 1) we identified a mutation in the ICOS gene as the first genetic cause of autosomal recessive CVID 2) we identified the first non-HLA linkage for autosomal dominant CVID, and 3) we identified a new linkage for autosomal recessive neutropenia. I am collaborating with P. Henthorn (U. Penn) and J. Fyfe (Michigan State U.) on a linkage study of a large canine pedigree that has multiple occurrences of three autosomal recessive traits. The highlight for the past 12 months is that we identified a linkage for a vitamin B12 malabsorption trait in a region that is not syntenic to human chromosome 10, which contains the gene for cubilin, which is mutated in some human cases of vitamin B12 malabsorption. I am collaborating with B. Nemesure (SUNY Stony Brook) and J. F. Hejtmancik (NEI) on a linkage study of glaucoma in Barbados. In the past 12 months we identified 2 loci of interest that have not been seen as susceptibility loci in previous studies limited to mostly Caucasian populations. I continue to collaborate with M. Mennoti-Raymond and W. J. Murphy (NCI) on construction of feline maps for use in genetics. In the past 12 months we constructed a new radiation hybrid map. Richard Desper and I are collaborating with J. Khan (NICHD) on a new method of tumor classification from microarray data. Our new method is based on a phylogenetic approach, and is particularly well-suited to data that has more than 2 categories, and data including samples that do not fit in any category. Richard Desper and I collaborated with Q. Huang (New York Eye and Ear) on a study of comparative genomic hybridization data from head and neck tumors.